A survey on periocular biometrics research

Alonso‐Fernandez, Fernando; Bigün, Josef

doi:10.1016/j.patrec.2015.08.026

Cited by 108 publications

(68 citation statements)

References 67 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…As baseline descriptors, we employ the most widely used features in periocular research [1]: LBP, HOG, and SIFT. In HOG and LBP, the image is decomposed into 8×8 non-overlapped regions.…”

Section: A Baseline Systemsmentioning

confidence: 99%

Periocular Recognition Using CNN Features Off-the-Shelf

Hernandez‐Diaz

Alonso‐Fernandez

Bigün

2018

2018 International Conference of the Biometrics Special Interest Group (BIOSIG)

Self Cite

View full text Add to dashboard Cite

Periocular refers to the region around the eye, including sclera, eyelids, lashes, brows and skin. With a surprisingly high discrimination ability, it is the ocular modality requiring the least constrained acquisition. Here, we apply existing pre-trained architectures, proposed in the context of the ImageNet Large Scale Visual Recognition Challenge, to the task of periocular recognition. These have proven to be very successful for many other computer vision tasks apart from the detection and classification tasks for which they were designed. Experiments are done with a database of periocular images captured with a digital camera. We demonstrate that these offthe-shelf CNN features can effectively recognize individuals based on periocular images, despite being trained to classify generic objects. Compared against reference periocular features, they show an EER reduction of up to ∼40%, with the fusion of CNN and traditional features providing additional improvements.

show abstract

“…As baseline descriptors, we employ the most widely used features in periocular research [1]: LBP, HOG, and SIFT. In HOG and LBP, the image is decomposed into 8×8 non-overlapped regions.…”

Section: A Baseline Systemsmentioning

confidence: 99%

Periocular Recognition Using CNN Features Off-the-Shelf

Hernandez‐Diaz

Alonso‐Fernandez

Bigün

2018

2018 International Conference of the Biometrics Special Interest Group (BIOSIG)

Self Cite

View full text Add to dashboard Cite

show abstract

“…Some researchers manually segment the periocular region by taking advantage of the ground truth eye centers provided with the database as reference points [8] [9]. There are three main techniques usually employed in literature for automatic detection and/or segmentation of the periocular region [44]. The first method is to initially detect the face with a face detector such as the Viola-Jones (VJ) algorithm [45].…”

Section: Periocular Image Segmentation and Preprocessing Methodsmentioning

confidence: 99%

“…The global approach extracts properties describing the whole region of interest (ROI) and performs global image analysis while the local approach first selects crucial locations, tagged key points, in the image. The properties in the neighborhood pixels of such key points are extracted based on some local analysis on them [44]. Other feature extraction methods which do not fall within these two broad groups are discussed in a separate category in this section.…”

Section: Periocular Image Representation Methodsmentioning

confidence: 99%

Survey of Periocular Recognition Techniques

Badejo¹,

Akinrinmade²,

Adetiba³

2019

JESTR

View full text Add to dashboard Cite

Periocular recognition has evolved over the years and has been shown to possess discriminative features for personal identification either as a stand-alone trait or when fused with other modalities such as face and iris, especially in unconstrained scenarios. It has a number of advantages, first being the region could be easily cropped from existing face images. Secondly, unlike the iris, its capturing process is less intrusive. This area of the face is easily captured with ease from the subject, for example by surveillance cameras. Thirdly, in crime scenes where masks hide the face or in cases where the subject's face is covered due to religious or cultural beliefs, the periocular region could still be captured providing useful details, that is, it is robust to face occlusion and is least affected by expression change. This survey relates the various techniques employed for periocular recognition at different stages: segmentation approaches, image preprocessing methods, feature extraction and matching algorithms. This survey is meant to facilitate a quick grasp of the development in this area for interested students, researchers as well as enthusiasts in the field of biometrics or any related application area. Information about the various databases used for performance evaluation of these techniques as well as the performance indicators is also provided.

show abstract

“…Several efforts have been made on periocular recognition, which have been reviewed and discussed in [23,24,25]. Most of the existing techniques for periocular recognition use handcrafted features.…”

Section: Introductionmentioning

confidence: 99%

ConvSRC: SmartPhone-based periocular recognition using deep convolutional neural network and sparsity augmented collaborative representation

Alahmadi

Hussain

Aboalsamh

et al. 2020

IFS

View full text Add to dashboard Cite

A B S T R A C TSmartphone based periocular recognition has gained significant attention from biometric research community because of the limitations of biometric modalities like face, iris etc. Most of the existing methods for periocular recognition employ hand-crafted features. Recently, learning based image representation techniques like deep Convolutional Neural Network (CNN) have shown outstanding performance in many visual recognition tasks. CNN needs a huge volume of data for its learning, but for periocular recognition only limited amount of data is available. The solution is to use CNN pre-trained on the dataset from the related domain, in this case the challenge is to extract efficiently the discriminative features. Using a pertained CNN model (VGG-Net), we propose a simple, efficient and compact image representation technique that takes into account the wealth of information and sparsity existing in the activations of the convolutional layers and employs principle component analysis. For recognition, we use an efficient and robust Sparse Augmented Collaborative Representation based Classification (SA-CRC) technique. For thorough evaluation of ConvSRC (the proposed system), experiments were carried out on the VISOB challenging database which was presented for periocular recognition competition in ICIP2016. The obtained results show the superiority of ConvSRC over the state-of-the-art methods; it obtains a GMR of more than 99% at FMR = 10 -3 and outperforms the first winner of ICIP2016 challenge by 10%.

show abstract

A survey on periocular biometrics research

Cited by 108 publications

References 67 publications

Periocular Recognition Using CNN Features Off-the-Shelf

Periocular Recognition Using CNN Features Off-the-Shelf

Survey of Periocular Recognition Techniques

ConvSRC: SmartPhone-based periocular recognition using deep convolutional neural network and sparsity augmented collaborative representation

Contact Info

Product

Resources

About